from sklearn_benchmarks.reporting.hp_match import HpMatchReporting
import pandas as pd
pd.set_option('display.max_colwidth', None)
pd.set_option('display.max_columns', None)
pd.set_option('display.max_rows', None)

scikit-learn vs. scikit-learn-intelex (Intel® oneAPI)

reporting = HpMatchReporting(other_library="sklearnex", config="config.yml", log_scale=True)
reporting.make_report()

We assume here there is a perfect match between the hyperparameters of both librairies. For a given set of parameters and a given dataset, we compute the speedup time scikit-learn / time sklearnex. For instance, a speedup of 2 means that sklearnex is twice as fast as scikit-learn for a given set of parameters and a given dataset.

KNeighborsClassifier_brute_force

scikit-learn-intelex (2021.20210705.191215) vs. scikit-learn (1.0.dev0)

Speedup barplots

All estimators share the following parameters: algorithm=brute.

Mismatches between validation scores

WARNING! Mismatch between validation scores for 18 predictions (75.0%).
The observed differences can be found in the diff_accuracy_scores column.
The chosen difference threshold is 0.001 for accuracy_score.
See details in the dataframe below.

	estimator	function	diff_accuracy_scores	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	n_iter	iteration_throughput	latency	algorithm	n_jobs	n_neighbors	accuracy_score_sklearn	mean_duration_sklearnex	std_duration_sklearnex	accuracy_score_sklearnex	speedup	std_speedup
1	KNeighborsClassifier_brute_force	predict	0.048	100000	1000	100	2.021145	0.040212	NaN	0.000396	0.002021	brute	-1	1	0.629	0.409500	0.010805	0.677	4.935642	4.937360
2	KNeighborsClassifier_brute_force	predict	1.000	100000	1	100	0.027088	0.002983	NaN	0.000030	0.027088	brute	-1	1	0.000	0.010904	0.000302	1.000	2.484203	2.485158
4	KNeighborsClassifier_brute_force	predict	0.011	100000	1000	100	2.993528	0.037443	NaN	0.000267	0.002994	brute	-1	5	0.688	0.405776	0.008452	0.677	7.377297	7.378897
5	KNeighborsClassifier_brute_force	predict	1.000	100000	1	100	0.027682	0.003318	NaN	0.000029	0.027682	brute	-1	5	0.000	0.011016	0.000442	1.000	2.512927	2.514946
7	KNeighborsClassifier_brute_force	predict	0.063	100000	1000	100	2.372928	0.019051	NaN	0.000337	0.002373	brute	1	100	0.797	0.410192	0.007477	0.734	5.784922	5.785883
8	KNeighborsClassifier_brute_force	predict	1.000	100000	1	100	0.022579	0.001141	NaN	0.000035	0.022579	brute	1	100	0.000	0.011469	0.000371	1.000	1.968671	1.969700
10	KNeighborsClassifier_brute_force	predict	0.008	100000	1000	100	3.005309	0.038994	NaN	0.000266	0.003005	brute	-1	100	0.797	0.457185	0.006563	0.805	6.573508	6.574186
11	KNeighborsClassifier_brute_force	predict	1.000	100000	1	100	0.027082	0.002393	NaN	0.000030	0.027082	brute	-1	100	0.000	0.010991	0.000461	1.000	2.463966	2.466131
13	KNeighborsClassifier_brute_force	predict	0.046	100000	1000	100	2.371938	0.016693	NaN	0.000337	0.002372	brute	1	5	0.688	0.403940	0.002088	0.734	5.872005	5.872084
14	KNeighborsClassifier_brute_force	predict	1.000	100000	1	100	0.023610	0.001646	NaN	0.000034	0.023610	brute	1	5	0.000	0.011638	0.001254	1.000	2.028690	2.040440
16	KNeighborsClassifier_brute_force	predict	0.176	100000	1000	100	1.402930	0.008560	NaN	0.000570	0.001403	brute	1	1	0.629	0.453800	0.009094	0.805	3.091512	3.092133
17	KNeighborsClassifier_brute_force	predict	1.000	100000	1	100	0.022103	0.000962	NaN	0.000036	0.022103	brute	1	1	0.000	0.010854	0.000373	1.000	2.036351	2.037551
19	KNeighborsClassifier_brute_force	predict	0.019	100000	1000	2	1.726828	0.024760	NaN	0.000009	0.001727	brute	-1	1	0.951	0.085150	0.002206	0.970	20.279723	20.286526
22	KNeighborsClassifier_brute_force	predict	0.001	100000	1000	2	2.755071	0.023094	NaN	0.000006	0.002755	brute	-1	5	0.971	0.085879	0.003378	0.970	32.080801	32.105605
25	KNeighborsClassifier_brute_force	predict	0.009	100000	1000	2	2.214361	0.009550	NaN	0.000007	0.002214	brute	1	100	0.969	0.085842	0.001648	0.978	25.795911	25.800662
28	KNeighborsClassifier_brute_force	predict	0.006	100000	1000	2	2.777270	0.022325	NaN	0.000006	0.002777	brute	-1	100	0.969	0.139361	0.004774	0.975	19.928533	19.940225
31	KNeighborsClassifier_brute_force	predict	0.007	100000	1000	2	2.205840	0.012963	NaN	0.000007	0.002206	brute	1	5	0.971	0.086497	0.002374	0.978	25.501951	25.511556
34	KNeighborsClassifier_brute_force	predict	0.024	100000	1000	2	1.171084	0.005302	NaN	0.000014	0.001171	brute	1	1	0.951	0.136675	0.003830	0.975	8.568405	8.571768

Raw results

fit

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	n_jobs	n_neighbors	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup	sklearn_profiling	sklearnex_profiling
0	KNeighborsClassifier_brute_force	fit	100000	100000	100	0.013	0.001	6.204	0.0	-1	1	0.051	0.002	0.252	0.252
3	KNeighborsClassifier_brute_force	fit	100000	100000	100	0.013	0.001	6.300	0.0	-1	5	0.052	0.001	0.242	0.242
6	KNeighborsClassifier_brute_force	fit	100000	100000	100	0.014	0.002	5.884	0.0	1	100	0.051	0.000	0.266	0.266
9	KNeighborsClassifier_brute_force	fit	100000	100000	100	0.013	0.000	6.396	0.0	-1	100	0.054	0.001	0.233	0.233
12	KNeighborsClassifier_brute_force	fit	100000	100000	100	0.013	0.000	6.351	0.0	1	5	0.052	0.002	0.242	0.243
15	KNeighborsClassifier_brute_force	fit	100000	100000	100	0.013	0.000	6.226	0.0	1	1	0.053	0.002	0.240	0.241
18	KNeighborsClassifier_brute_force	fit	100000	100000	2	0.005	0.000	0.316	0.0	-1	1	0.009	0.001	0.552	0.553
21	KNeighborsClassifier_brute_force	fit	100000	100000	2	0.005	0.000	0.325	0.0	-1	5	0.009	0.000	0.534	0.534
24	KNeighborsClassifier_brute_force	fit	100000	100000	2	0.005	0.000	0.320	0.0	1	100	0.010	0.002	0.513	0.520
27	KNeighborsClassifier_brute_force	fit	100000	100000	2	0.005	0.000	0.324	0.0	-1	100	0.009	0.000	0.538	0.538
30	KNeighborsClassifier_brute_force	fit	100000	100000	2	0.006	0.001	0.288	0.0	1	5	0.009	0.000	0.648	0.649
33	KNeighborsClassifier_brute_force	fit	100000	100000	2	0.005	0.001	0.315	0.0	1	1	0.009	0.000	0.565	0.566

predict

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	n_jobs	n_neighbors	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup	sklearn_profiling	sklearnex_profiling
1	KNeighborsClassifier_brute_force	predict	100000	1000	100	2.021	0.040	0.000	0.002	-1	1	0.409	0.011	4.936	4.937
2	KNeighborsClassifier_brute_force	predict	100000	1	100	0.027	0.003	0.000	0.027	-1	1	0.011	0.000	2.484	2.485
4	KNeighborsClassifier_brute_force	predict	100000	1000	100	2.994	0.037	0.000	0.003	-1	5	0.406	0.008	7.377	7.379
5	KNeighborsClassifier_brute_force	predict	100000	1	100	0.028	0.003	0.000	0.028	-1	5	0.011	0.000	2.513	2.515
7	KNeighborsClassifier_brute_force	predict	100000	1000	100	2.373	0.019	0.000	0.002	1	100	0.410	0.007	5.785	5.786
8	KNeighborsClassifier_brute_force	predict	100000	1	100	0.023	0.001	0.000	0.023	1	100	0.011	0.000	1.969	1.970
10	KNeighborsClassifier_brute_force	predict	100000	1000	100	3.005	0.039	0.000	0.003	-1	100	0.457	0.007	6.574	6.574
11	KNeighborsClassifier_brute_force	predict	100000	1	100	0.027	0.002	0.000	0.027	-1	100	0.011	0.000	2.464	2.466
13	KNeighborsClassifier_brute_force	predict	100000	1000	100	2.372	0.017	0.000	0.002	1	5	0.404	0.002	5.872	5.872
14	KNeighborsClassifier_brute_force	predict	100000	1	100	0.024	0.002	0.000	0.024	1	5	0.012	0.001	2.029	2.040
16	KNeighborsClassifier_brute_force	predict	100000	1000	100	1.403	0.009	0.001	0.001	1	1	0.454	0.009	3.092	3.092
17	KNeighborsClassifier_brute_force	predict	100000	1	100	0.022	0.001	0.000	0.022	1	1	0.011	0.000	2.036	2.038
19	KNeighborsClassifier_brute_force	predict	100000	1000	2	1.727	0.025	0.000	0.002	-1	1	0.085	0.002	20.280	20.287
20	KNeighborsClassifier_brute_force	predict	100000	1	2	0.011	0.006	0.000	0.011	-1	1	0.001	0.000	19.016	19.348
22	KNeighborsClassifier_brute_force	predict	100000	1000	2	2.755	0.023	0.000	0.003	-1	5	0.086	0.003	32.081	32.106
23	KNeighborsClassifier_brute_force	predict	100000	1	2	0.008	0.002	0.000	0.008	-1	5	0.001	0.000	12.701	12.951
25	KNeighborsClassifier_brute_force	predict	100000	1000	2	2.214	0.010	0.000	0.002	1	100	0.086	0.002	25.796	25.801
26	KNeighborsClassifier_brute_force	predict	100000	1	2	0.003	0.000	0.000	0.003	1	100	0.001	0.000	5.241	5.324
28	KNeighborsClassifier_brute_force	predict	100000	1000	2	2.777	0.022	0.000	0.003	-1	100	0.139	0.005	19.929	19.940
29	KNeighborsClassifier_brute_force	predict	100000	1	2	0.010	0.002	0.000	0.010	-1	100	0.001	0.000	15.222	15.424
31	KNeighborsClassifier_brute_force	predict	100000	1000	2	2.206	0.013	0.000	0.002	1	5	0.086	0.002	25.502	25.512
32	KNeighborsClassifier_brute_force	predict	100000	1	2	0.003	0.000	0.000	0.003	1	5	0.001	0.000	5.157	5.231
34	KNeighborsClassifier_brute_force	predict	100000	1000	2	1.171	0.005	0.000	0.001	1	1	0.137	0.004	8.568	8.572
35	KNeighborsClassifier_brute_force	predict	100000	1	2	0.002	0.000	0.000	0.002	1	1	0.001	0.000	2.980	3.020

KNeighborsClassifier_kd_tree

scikit-learn-intelex (2021.20210705.191215) vs. scikit-learn (1.0.dev0)

Speedup barplots

All estimators share the following parameters: algorithm=kd_tree.

Mismatches between validation scores

WARNING! Mismatch between validation scores for 12 predictions (50.0%).
The observed differences can be found in the diff_accuracy_scores column.
The chosen difference threshold is 0.001 for accuracy_score.
See details in the dataframe below.

	estimator	function	diff_accuracy_scores	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	n_iter	iteration_throughput	latency	algorithm	n_jobs	n_neighbors	accuracy_score_sklearn	mean_duration_sklearnex	std_duration_sklearnex	accuracy_score_sklearnex	speedup	std_speedup
1	KNeighborsClassifier_kd_tree	predict	0.009	1000000	1000	10	3.664288	0.021723	NaN	0.000022	0.003664	kd_tree	-1	100	0.907	0.140077	0.006822	0.898	26.159020	26.190022
4	KNeighborsClassifier_kd_tree	predict	0.009	1000000	1000	10	6.279532	0.022999	NaN	0.000013	0.006280	kd_tree	1	100	0.907	0.137837	0.002814	0.898	45.557605	45.567095
7	KNeighborsClassifier_kd_tree	predict	0.062	1000000	1000	10	1.038654	0.015005	NaN	0.000077	0.001039	kd_tree	1	1	0.868	0.721264	0.009439	0.930	1.440048	1.440171
10	KNeighborsClassifier_kd_tree	predict	0.019	1000000	1000	10	1.149265	0.013332	NaN	0.000070	0.001149	kd_tree	-1	5	0.912	0.247547	0.004057	0.931	4.642610	4.643233
13	KNeighborsClassifier_kd_tree	predict	0.063	1000000	1000	10	0.614735	0.005860	NaN	0.000130	0.000615	kd_tree	-1	1	0.868	0.248266	0.003278	0.931	2.476113	2.476329
16	KNeighborsClassifier_kd_tree	predict	0.018	1000000	1000	10	1.982398	0.011555	NaN	0.000040	0.001982	kd_tree	1	5	0.912	0.737868	0.015006	0.930	2.686656	2.687212
19	KNeighborsClassifier_kd_tree	predict	0.041	1000	1000	2	0.050027	0.009165	NaN	0.000320	0.000050	kd_tree	-1	100	0.910	0.000580	0.000095	0.951	86.247184	87.402205
22	KNeighborsClassifier_kd_tree	predict	0.041	1000	1000	2	0.042774	0.001215	NaN	0.000374	0.000043	kd_tree	1	100	0.910	0.000566	0.000113	0.951	75.541148	77.020639
25	KNeighborsClassifier_kd_tree	predict	0.044	1000	1000	2	0.028988	0.000969	NaN	0.000552	0.000029	kd_tree	1	1	0.948	0.005078	0.001437	0.904	5.708113	5.932131
28	KNeighborsClassifier_kd_tree	predict	0.009	1000	1000	2	0.035712	0.003421	NaN	0.000448	0.000036	kd_tree	-1	5	0.958	0.000877	0.000177	0.949	40.725710	41.545230
31	KNeighborsClassifier_kd_tree	predict	0.001	1000	1000	2	0.032502	0.001419	NaN	0.000492	0.000033	kd_tree	-1	1	0.948	0.000825	0.000112	0.949	39.373041	39.735831
34	KNeighborsClassifier_kd_tree	predict	0.054	1000	1000	2	0.031148	0.001660	NaN	0.000514	0.000031	kd_tree	1	5	0.958	0.004434	0.000252	0.904	7.025005	7.036332

Raw results

fit

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	n_jobs	n_neighbors	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup	sklearn_profiling	sklearnex_profiling
0	KNeighborsClassifier_kd_tree	fit	1000000	1000000	10	2.961	0.031	0.027	0.0	-1	100	0.795	0.018	3.726	3.727
3	KNeighborsClassifier_kd_tree	fit	1000000	1000000	10	4.671	0.035	0.017	0.0	1	100	0.780	0.018	5.988	5.989
6	KNeighborsClassifier_kd_tree	fit	1000000	1000000	10	4.691	0.051	0.017	0.0	1	1	0.788	0.014	5.952	5.953
9	KNeighborsClassifier_kd_tree	fit	1000000	1000000	10	4.718	0.066	0.017	0.0	-1	5	0.737	0.014	6.404	6.405
12	KNeighborsClassifier_kd_tree	fit	1000000	1000000	10	4.731	0.031	0.017	0.0	-1	1	0.777	0.017	6.093	6.094
15	KNeighborsClassifier_kd_tree	fit	1000000	1000000	10	4.730	0.048	0.017	0.0	1	5	0.749	0.018	6.312	6.314
18	KNeighborsClassifier_kd_tree	fit	1000	1000	2	0.001	0.000	0.021	0.0	-1	100	0.004	0.002	0.213	0.237
21	KNeighborsClassifier_kd_tree	fit	1000	1000	2	0.001	0.000	0.024	0.0	1	100	0.004	0.003	0.164	0.202
24	KNeighborsClassifier_kd_tree	fit	1000	1000	2	0.001	0.000	0.024	0.0	1	1	0.001	0.000	0.629	0.639
27	KNeighborsClassifier_kd_tree	fit	1000	1000	2	0.001	0.000	0.023	0.0	-1	5	0.001	0.000	0.682	0.685
30	KNeighborsClassifier_kd_tree	fit	1000	1000	2	0.001	0.000	0.023	0.0	-1	1	0.001	0.000	0.692	0.696
33	KNeighborsClassifier_kd_tree	fit	1000	1000	2	0.001	0.000	0.023	0.0	1	5	0.001	0.000	0.662	0.669

predict

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	n_jobs	n_neighbors	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup	sklearn_profiling	sklearnex_profiling
1	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	3.664	0.022	0.000	0.004	-1	100	0.140	0.007	26.159	26.190
2	KNeighborsClassifier_kd_tree	predict	1000000	1	10	0.011	0.001	0.000	0.011	-1	100	0.001	0.000	17.632	18.646
4	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	6.280	0.023	0.000	0.006	1	100	0.138	0.003	45.558	45.567
5	KNeighborsClassifier_kd_tree	predict	1000000	1	10	0.007	0.001	0.000	0.007	1	100	0.001	0.000	12.177	13.095
7	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	1.039	0.015	0.000	0.001	1	1	0.721	0.009	1.440	1.440
8	KNeighborsClassifier_kd_tree	predict	1000000	1	10	0.002	0.001	0.000	0.002	1	1	0.002	0.000	0.980	1.011
10	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	1.149	0.013	0.000	0.001	-1	5	0.248	0.004	4.643	4.643
11	KNeighborsClassifier_kd_tree	predict	1000000	1	10	0.005	0.001	0.000	0.005	-1	5	0.001	0.000	6.273	6.610
13	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	0.615	0.006	0.000	0.001	-1	1	0.248	0.003	2.476	2.476
14	KNeighborsClassifier_kd_tree	predict	1000000	1	10	0.004	0.000	0.000	0.004	-1	1	0.001	0.000	5.364	5.662
16	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	1.982	0.012	0.000	0.002	1	5	0.738	0.015	2.687	2.687
17	KNeighborsClassifier_kd_tree	predict	1000000	1	10	0.002	0.001	0.000	0.002	1	5	0.002	0.001	1.415	1.484
19	KNeighborsClassifier_kd_tree	predict	1000	1000	2	0.050	0.009	0.000	0.000	-1	100	0.001	0.000	86.247	87.402
20	KNeighborsClassifier_kd_tree	predict	1000	1	2	0.003	0.000	0.000	0.003	-1	100	0.000	0.000	18.175	19.762
22	KNeighborsClassifier_kd_tree	predict	1000	1000	2	0.043	0.001	0.000	0.000	1	100	0.001	0.000	75.541	77.021
23	KNeighborsClassifier_kd_tree	predict	1000	1	2	0.001	0.000	0.000	0.001	1	100	0.000	0.000	4.087	4.895
25	KNeighborsClassifier_kd_tree	predict	1000	1000	2	0.029	0.001	0.001	0.000	1	1	0.005	0.001	5.708	5.932
26	KNeighborsClassifier_kd_tree	predict	1000	1	2	0.001	0.000	0.000	0.001	1	1	0.000	0.000	4.565	5.252
28	KNeighborsClassifier_kd_tree	predict	1000	1000	2	0.036	0.003	0.000	0.000	-1	5	0.001	0.000	40.726	41.545
29	KNeighborsClassifier_kd_tree	predict	1000	1	2	0.003	0.000	0.000	0.003	-1	5	0.000	0.000	17.236	19.242
31	KNeighborsClassifier_kd_tree	predict	1000	1000	2	0.033	0.001	0.000	0.000	-1	1	0.001	0.000	39.373	39.736
32	KNeighborsClassifier_kd_tree	predict	1000	1	2	0.003	0.000	0.000	0.003	-1	1	0.000	0.000	15.979	17.362
34	KNeighborsClassifier_kd_tree	predict	1000	1000	2	0.031	0.002	0.001	0.000	1	5	0.004	0.000	7.025	7.036
35	KNeighborsClassifier_kd_tree	predict	1000	1	2	0.001	0.000	0.000	0.001	1	5	0.000	0.000	4.967	5.383

KMeans_tall

scikit-learn-intelex (2021.20210705.191215) vs. scikit-learn (1.0.dev0)

Speedup barplots

All estimators share the following parameters: algorithm=full, n_clusters=3, max_iter=30, n_init=1, tol=1e-16.

Raw results

fit

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	n_iter	iteration_throughput	latency	init	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup	sklearn_profiling	sklearnex_profiling
0	KMeans_tall	fit	1000000	1000000	2	0.664	0.016	30	0.024	0.0	k-means++	0.289	0.005	2.299	2.299
3	KMeans_tall	fit	1000000	1000000	2	0.557	0.007	30	0.029	0.0	random	0.334	0.014	1.667	1.668
6	KMeans_tall	fit	1000000	1000000	100	7.069	0.049	30	0.113	0.0	k-means++	3.777	0.031	1.872	1.872
9	KMeans_tall	fit	1000000	1000000	100	6.572	0.042	30	0.122	0.0	random	3.977	0.020	1.652	1.652

predict

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	n_iter	iteration_throughput	latency	init	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup	sklearn_profiling	sklearnex_profiling
1	KMeans_tall	predict	1000000	1000	2	0.002	0.0	30	0.009	0.000	k-means++	0.0	0.0	7.471	7.820
2	KMeans_tall	predict	1000000	1	2	0.002	0.0	30	0.000	0.002	k-means++	0.0	0.0	9.510	10.174
4	KMeans_tall	predict	1000000	1000	2	0.002	0.0	30	0.009	0.000	random	0.0	0.0	7.478	7.983
5	KMeans_tall	predict	1000000	1	2	0.002	0.0	30	0.000	0.002	random	0.0	0.0	10.266	11.132
7	KMeans_tall	predict	1000000	1000	100	0.002	0.0	30	0.386	0.000	k-means++	0.0	0.0	5.840	7.115
8	KMeans_tall	predict	1000000	1	100	0.002	0.0	30	0.000	0.002	k-means++	0.0	0.0	8.749	9.470
10	KMeans_tall	predict	1000000	1000	100	0.002	0.0	30	0.389	0.000	random	0.0	0.0	6.537	6.913
11	KMeans_tall	predict	1000000	1	100	0.002	0.0	30	0.000	0.002	random	0.0	0.0	8.691	9.339

KMeans_short

scikit-learn-intelex (2021.20210705.191215) vs. scikit-learn (1.0.dev0)

Speedup barplots

All estimators share the following parameters: algorithm=full, n_clusters=300, max_iter=20, n_init=1, tol=1e-16.

Mismatches between validation scores

WARNING! Mismatch between validation scores for 3 predictions (37.5%).
The observed differences can be found in the diff_adjusted_rand_scores column.
The chosen difference threshold is 0.001 for adjusted_rand_score.
See details in the dataframe below.

	estimator	function	diff_adjusted_rand_scores	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	n_iter	iteration_throughput	latency	algorithm	init	max_iter	n_clusters	n_init	tol	adjusted_rand_score_sklearn	mean_duration_sklearnex	std_duration_sklearnex	adjusted_rand_score_sklearnex	speedup	std_speedup
4	KMeans_short	predict	0.004085	10000	1000	2	0.002223	0.000336	20	0.007196	0.000002	full	k-means++	20	300	1	1.000000e-16	0.003724	0.000670	0.000212	-0.000361	3.317051	3.478682
7	KMeans_short	predict	0.019341	10000	1000	100	0.003351	0.000309	20	0.238743	0.000003	full	random	20	300	1	1.000000e-16	0.348183	0.001483	0.000182	0.328841	2.258812	2.275814
10	KMeans_short	predict	0.015990	10000	1000	100	0.003308	0.000295	20	0.241874	0.000003	full	k-means++	20	300	1	1.000000e-16	0.303254	0.001558	0.000258	0.287264	2.123517	2.152343

Raw results

fit

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	n_iter	iteration_throughput	latency	init	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup	sklearn_profiling	sklearnex_profiling
0	KMeans_short	fit	10000	10000	2	0.095	0.006	20	0.002	0.0	random	0.049	0.001	1.949	1.949
3	KMeans_short	fit	10000	10000	2	0.294	0.004	20	0.001	0.0	k-means++	0.131	0.006	2.234	2.236
6	KMeans_short	fit	10000	10000	100	0.318	0.005	20	0.025	0.0	random	0.251	0.004	1.270	1.270
9	KMeans_short	fit	10000	10000	100	1.081	0.019	20	0.007	0.0	k-means++	0.600	0.006	1.803	1.803

predict

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	n_iter	iteration_throughput	latency	init	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup	sklearn_profiling	sklearnex_profiling
1	KMeans_short	predict	10000	1000	2	0.002	0.000	20	0.008	0.000	random	0.001	0.000	3.050	3.126
2	KMeans_short	predict	10000	1	2	0.002	0.000	20	0.000	0.002	random	0.000	0.000	8.627	9.882
4	KMeans_short	predict	10000	1000	2	0.002	0.000	20	0.007	0.000	k-means++	0.001	0.000	3.317	3.479
5	KMeans_short	predict	10000	1	2	0.002	0.000	20	0.000	0.002	k-means++	0.000	0.000	9.097	9.687
7	KMeans_short	predict	10000	1000	100	0.003	0.000	20	0.239	0.000	random	0.001	0.000	2.259	2.276
8	KMeans_short	predict	10000	1	100	0.003	0.003	20	0.000	0.003	random	0.001	0.002	3.854	8.939
10	KMeans_short	predict	10000	1000	100	0.003	0.000	20	0.242	0.000	k-means++	0.002	0.000	2.124	2.152
11	KMeans_short	predict	10000	1	100	0.002	0.000	20	0.000	0.002	k-means++	0.000	0.000	8.704	9.285

LogisticRegression

scikit-learn-intelex (2021.20210705.191215) vs. scikit-learn (1.0.dev0)

Speedup barplots

All estimators share the following parameters: penalty=l2, dual=False, tol=0.0001, C=1.0, fit_intercept=True, intercept_scaling=1, class_weight=nan, random_state=nan, solver=lbfgs, max_iter=100, multi_class=auto, verbose=0, warm_start=False, n_jobs=nan, l1_ratio=nan.

Mismatches between validation scores

WARNING! Mismatch between validation scores for 3 predictions (75.0%).
The observed differences can be found in the diff_accuracy_scores column.
The chosen difference threshold is 0.001 for accuracy_score.
See details in the dataframe below.

	estimator	function	diff_accuracy_scores	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	n_iter	iteration_throughput	latency	C	class_weight	dual	fit_intercept	intercept_scaling	l1_ratio	max_iter	multi_class	n_jobs	penalty	random_state	solver	tol	verbose	warm_start	accuracy_score_sklearn	mean_duration_sklearnex	std_duration_sklearnex	accuracy_score_sklearnex	speedup	std_speedup
1	LogisticRegression	predict	0.006	1000000	1000	100	0.000354	0.000095	[17]	2.260383	3.539223e-07	1.0	NaN	False	True	1	NaN	100	auto	NaN	l2	NaN	lbfgs	0.0001	0	False	0.611	0.000400	0.000099	0.605	0.883737	0.910129
2	LogisticRegression	predict	1.000	1000000	1	100	0.000094	0.000048	[17]	0.008556	9.350320e-05	1.0	NaN	False	True	1	NaN	100	auto	NaN	l2	NaN	lbfgs	0.0001	0	False	0.000	0.000251	0.000088	1.000	0.372970	0.395276
4	LogisticRegression	predict	0.040	1000	100	10000	0.002309	0.000156	[26]	3.465245	2.308639e-05	1.0	NaN	False	True	1	NaN	100	auto	NaN	l2	NaN	lbfgs	0.0001	0	False	0.280	0.004128	0.000287	0.240	0.559228	0.560579

Raw results

fit

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	n_iter	iteration_throughput	latency	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup	sklearn_profiling	sklearnex_profiling
0	LogisticRegression	fit	1000000	1000000	100	15.048	0.111	[17]	0.053	0.000	2.698	0.008	5.577	5.577
3	LogisticRegression	fit	1000	1000	10000	1.236	0.031	[26]	0.065	0.001	1.070	0.030	1.155	1.156

predict

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	n_iter	iteration_throughput	latency	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup	sklearn_profiling	sklearnex_profiling
1	LogisticRegression	predict	1000000	1000	100	0.000	0.0	[17]	2.260	0.0	0.000	0.0	0.884	0.910
2	LogisticRegression	predict	1000000	1	100	0.000	0.0	[17]	0.009	0.0	0.000	0.0	0.373	0.395
4	LogisticRegression	predict	1000	100	10000	0.002	0.0	[26]	3.465	0.0	0.004	0.0	0.559	0.561
5	LogisticRegression	predict	1000	1	10000	0.000	0.0	[26]	0.631	0.0	0.001	0.0	0.133	0.138

Ridge

scikit-learn-intelex (2021.20210705.191215) vs. scikit-learn (1.0.dev0)

Speedup barplots

All estimators share the following parameters: alpha=1.0, fit_intercept=True, normalize=deprecated, copy_X=True, max_iter=nan, tol=0.001, solver=auto, random_state=nan.

Mismatches between validation scores

WARNING! Mismatch between validation scores for 1 prediction (25.0%).
The observed differences can be found in the diff_r2_scores column.
The chosen difference threshold is 0.001 for r2_score.
See details in the dataframe below.

	estimator	function	diff_r2_scores	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	n_iter	iteration_throughput	latency	alpha	copy_X	fit_intercept	max_iter	normalize	random_state	solver	tol	r2_score_sklearn	mean_duration_sklearnex	std_duration_sklearnex	r2_score_sklearnex	speedup	std_speedup
1	Ridge	predict	0.010925	1000	1000	10000	0.01204	0.000956	NaN	6.644538	0.000012	1.0	True	True	NaN	deprecated	NaN	auto	0.001	0.086907	0.019506	0.001049	0.097831	0.617246	0.618137

Raw results

fit

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup	sklearn_profiling	sklearnex_profiling
0	Ridge	fit	1000	1000	10000	0.271	0.006	0.296	0.0	0.300	0.005	0.904	0.904
3	Ridge	fit	1000000	1000000	100	1.252	0.052	0.639	0.0	0.341	0.004	3.669	3.669

predict

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	mean_duration_sklearnex	std_duration_sklearnex	speedup	std_speedup	sklearn_profiling	sklearnex_profiling
1	Ridge	predict	1000	1000	10000	0.012	0.001	6.645	0.0	0.02	0.001	0.617	0.618
2	Ridge	predict	1000	1	10000	0.000	0.000	0.774	0.0	0.00	0.000	0.643	0.694
4	Ridge	predict	1000000	1000	100	0.000	0.000	4.493	0.0	0.00	0.000	0.667	0.702
5	Ridge	predict	1000000	1	100	0.000	0.000	0.009	0.0	0.00	0.000	0.601	0.659